Wyniki wyszukiwania - BazTech

1

Effectiveness of Fast Fourier Transform implementations on GPU and CPU

Puchała D., Stokfiszewski K., Szczepaniak B., Yatsymirskyy M.

Przegląd Elektrotechniczny

|

2016

|

R. 92, nr 7

69--71

EN

In this paper, we present the results of comparison of the effectiveness of selected variants of radix-2 Fast Fourier Transform (FFT) algorithms implemented on both Graphics (GPU) and Central (CPU) Processing Units. The considered algorithms differ in memory consumption and the arrangement of data-flow paths which affects the global memory coalescing and cache memory exploitation. The obtained results allow to indicate the variants of FFT algorithms which are best suited for GPU and CPU architectures, to confirm the advisability of GPU oriented calculations of FFT and to formulate a guideline for implementations of fast algorithms of various linear transforms.

XX

W niniejszej pracy przedstawiono wyniki porównania efektywności wybranych wariantów algorytmów szybkiej transformaty Fouriera (FFT) typu radix-2 realizowanych zarówno dla procesorów graficznych (GPU) jak i typowych jednostek centralnych (CPU). Rozważane algorytmy różnią się zapotrzebowaniem pamięciowym oraz postaciami grafów przepływu danych, które mają wpływ na spójność wykorzystania pamięci globalnej oraz pamięci cache jednostek GPU i CPU. Uzyskane wyniki pozwalają na wskazanie wariantów algorytmów FFT, które są najlepiej dostosowane dla architektur GPU i CPU, pozwalają też potwierdzić celowość realizacji implementacji FFT zorientowanych na wykorzystanie jednostek GPU, a także sformułować ogólne wytyczne dla implementacji zorientowanych na wykorzystanie jednostek GPU algorytmów szybkich przekształceń liniowych.

2

Time effectiveness optimization of cross correlation methods for ocr systems with the use of graphics processing units

Wieloch K., Stokfiszewski K.

Journal of Applied Computer Science

|

2015

|

Vol. 23, nr 2

79--100

EN

In this paper authors propose a time optimization of fast normalized cross corrlation methods for image processing and optical character recognition with the use of parallel computing techniques realized on graphics processing units (GPU). It is shown that suitable modification of the wellknown formulas and their parallel implementation on graphics processing units may substantially accelerate computing time without any change of the quality of results. The performed research include comparative analysis of time efficiency of the developed methods with respect to their standard sequential implementations.

3

Image filtering with fast parametrized biorthogonal transforms implemented on a new GUI research aid system

Puchała D., Stasiak B., Stokfiszewski K., Yatsymirskyy M.

Journal of Applied Computer Science

|

2013

|

Vol. 21, nr 2

97--115

EN

In this paper the authors show that fast parametrized biorthogonal transforms (FPBT) are well suited for adaptive generalized Wiener image ﬁltering. Research results are obtained with a use of a new graphical user interface system for implementing various fast adaptive techniques, designed, implemented and published by the authors as a part of a project Innovative Economy Programme 2007-2013 „Platforma Informatyczna TEWI”.

4

Low-complexity approximation of 8-point discrete cosine transform for image compression

Puchała D., Stokfiszewski K.

Journal of Applied Computer Science

|

2012

|

Vol. 20, nr 2

107-117

EN

In this paper the authors propose a new low-complexity approximation of 8-point discrete cosine transform (DCT) that requires 18 additions and two bit-shift operations. It is shown that the proposed transform outperforms significantly the known transform of the same computational complexity when applied to a JPEG compression stream in practical cases of encoding and decoding of still images. As such, the proposed transform can be effectively used in any practical applications where significant limitations exist regarding the computational capabilities coding and / or decoding devices, i.e. mobile devices or industrial imaging devices.